A method and apparatus for determining a class reference data record for classifying documents of value includes creating a class reference image using a multiplicity of reference images of already classified documents of value in the same class; and creating a class reference data record having at least one class reference parameter using the class reference image. The pixel and intensity values of the respective pixel in the class reference image are a function of the pixel values of the relevant pixels in the multiplicity of reference images of already classified documents of value in the same class. The method includes determining at least one quantitative property of a document of value to be classified; and classifying the document of value to be classified on the basis of a comparison between the quantitative property of the document of value to be classified and the class reference data record.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. At a banknote evaluation machine comprising a sensor unit with at least one imaging optical sensor for creating digital images of bank notes, a processor and a memory, a method for determining a class reference data record for the classification of value documents, in particular bank notes, the method comprising the following steps: the banknote evaluation machine creating a class reference image by means of a plurality of reference images of already classified value documents of same class, the plurality of reference images being created using the imaging optical sensor; and creating, by using the processor, a class reference data record having at least one class reference parameter with the aid of a user-defined region having at least one optical feature within the class reference image, the class reference parameter being coupled to at least one quantitatively determined property of the user-defined region, the class reference parameter being created by forming a ratio of pixel values from at least two different locations on a value document, the ratio being formed from a pixel value averaged over a specified region of an infrared-absorbing section of the value document to a pixel value averaged over a specified region of an infrared-transparent section of the value document; wherein the pixel values of a respective pixel of the class reference image are at least one of the average pixel values or median pixel values of the corresponding pixels of the plurality of reference images of already classified value documents of the same class, wherein each class reference parameter has assigned thereto a tolerance range, the tolerance range for each class reference parameter being variable and being dependent on the quantitatively determined property of the user-defined region; and wherein the class reference image of a class of value documents is represented, for analysis, on a display unit by means of a graphical user interface.
A banknote evaluation machine classifies banknotes by creating a "class reference image" from multiple images of known, correctly classified banknotes of the same type, acquired using an optical sensor. A "class reference data record" is created using a processor, based on a user-defined region in the class reference image that contains optical features. A class reference parameter, linked to a quantitative property of this region, is created by calculating the ratio of averaged pixel values from an infrared-absorbing area to an infrared-transparent area on the banknote. Each pixel in the class reference image is either the average or median of the corresponding pixels from the reference images. Each class reference parameter is assigned a variable tolerance range dependent on the region's quantitative property. The class reference image is displayed on a screen via a graphical user interface for analysis.
2. The method according to claim 1 , wherein in an adaptation process for each class to which a value document can be allocated there are created a class reference image and a class reference data record having at least one class reference parameter by means of a plurality of reference images of reference value documents of a respective class.
In an adaptation process, a banknote evaluation machine creates a class reference image and a class reference data record with at least one class reference parameter for each banknote class. This creation uses multiple reference images of known banknotes belonging to that specific class. This builds upon the method where a class reference image is created from images of known banknotes, a user-defined region selected in the image, and a class reference parameter calculated as a ratio of pixel values from infrared-absorbing and transparent regions. Each pixel in the class reference image is either the average or median of corresponding pixels from the reference images and a variable tolerance range is assigned to the class reference parameter.
3. The method according to claim 2 , wherein, during the adaptation process, the class reference image of a class of value documents is represented, for the analysis, on a display unit by means of a graphical user interface, wherein the graphical user interface is configured such that a user can participate in the creation of a class reference parameter of a class reference data record with the aid of the class reference image.
During the adaptation process, the class reference image for each banknote class is displayed on a screen using a graphical user interface. The user interface enables the user to participate in defining the class reference parameter of the class reference data record, using the visual information from the displayed class reference image. This builds upon the methods where a class reference image and a class reference data record with a class reference parameter are created using multiple reference images of known banknotes, a user-defined region selected in the image, and a class reference parameter calculated as a ratio of pixel values from infrared-absorbing and transparent regions with a variable tolerance range, with the class reference image created from average/median pixel values from corresponding pixels from reference images.
4. The method according to claim 1 , wherein the step of creating at least one class reference image by means of a plurality of reference images of already classified value documents of the same class preferably further comprises aligning the plurality of reference images of already classified value documents of the same class with respect to a common reference point.
When creating a class reference image from multiple reference images of known banknotes, the process includes aligning these reference images relative to a common reference point before further processing. This alignment improves the accuracy of the resulting class reference image. This builds upon the method where a class reference image is created from images of known banknotes, a user-defined region selected in the image, and a class reference parameter calculated as a ratio of pixel values from infrared-absorbing and transparent regions. Each pixel in the class reference image is either the average or median of corresponding pixels from the reference images and a variable tolerance range is assigned to the class reference parameter. The class reference image is displayed via a GUI.
5. The method according to claim 1 , wherein for some or all of the classes more than one class reference image is created by means of the plurality of reference images of already classified value documents of the same class and the class reference data record having at least one class reference parameter is created with the aid of more than one class reference image.
For some or all banknote classes, the system creates multiple class reference images using multiple reference images of already classified banknotes, and then builds a class reference data record with at least one class reference parameter based on these multiple class reference images. This allows for representing variations within the same class. This extends upon the method where a single class reference image is created from images of known banknotes, a user-defined region selected in the image, and a class reference parameter calculated as a ratio of pixel values from infrared-absorbing and transparent regions. Each pixel in the class reference image is either the average or median of corresponding pixels from the reference images and a variable tolerance range is assigned to the class reference parameter.
6. The method according to claim 1 , wherein the pixel values of a pixel of a class reference image are the minima of the corresponding pixel values of the corresponding pixels of the plurality of reference images of already classified value documents of the same class.
The pixel values of a pixel in a class reference image are determined by the minimum pixel value of the corresponding pixels in the multiple reference images of correctly classified banknotes. Instead of averages or medians, the lowest value is selected. This builds upon the method where a class reference image is created from images of known banknotes, a user-defined region selected in the image, and a class reference parameter calculated as a ratio of pixel values from infrared-absorbing and transparent regions. A variable tolerance range is assigned to the class reference parameter and the class reference image is displayed via a GUI.
7. The method according to claim 1 , wherein the pixel values of a pixel of the class reference image are the maxima of the corresponding pixel values of the corresponding pixels of the plurality of reference images of already classified value documents of the same class.
The pixel values of a pixel in a class reference image are determined by the maximum pixel value of the corresponding pixels in the multiple reference images of correctly classified banknotes. Instead of averages or medians, the highest value is selected. This builds upon the method where a class reference image is created from images of known banknotes, a user-defined region selected in the image, and a class reference parameter calculated as a ratio of pixel values from infrared-absorbing and transparent regions. A variable tolerance range is assigned to the class reference parameter and the class reference image is displayed via a GUI.
8. The method according to claim 1 , wherein the pixel values of a respective pixel of the class reference image are determined by the difference between the maximum pixel value and the minimum pixel value of the corresponding pixels of the plurality of reference images of already classified value documents of the same class.
The pixel values of a pixel in the class reference image are determined by calculating the difference between the maximum and minimum pixel values of the corresponding pixels in the multiple reference images of already classified banknotes. This difference represents the pixel value. This builds upon the method where a class reference image is created from images of known banknotes, a user-defined region selected in the image, and a class reference parameter calculated as a ratio of pixel values from infrared-absorbing and transparent regions. Each pixel in the class reference image is either the average or median of corresponding pixels from the reference images and a variable tolerance range is assigned to the class reference parameter, and a graphical user interface.
9. The method according to claim 1 , the method further comprising the following steps: determining at least one quantitative property of a value document to be classified; and classifying the value document to be classified on the basis of a comparison of the quantitative property of the value document to be classified with a class reference parameter of a class reference data record.
The method involves determining at least one quantitative property of a banknote to be classified and then classifying that banknote based on a comparison of the quantitative property with a class reference parameter from a pre-existing class reference data record. This builds upon the method where a class reference image is created from images of known banknotes, a user-defined region selected in the image, and a class reference parameter calculated as a ratio of pixel values from infrared-absorbing and transparent regions. Each pixel in the class reference image is either the average or median of corresponding pixels from the reference images and a variable tolerance range is assigned to the class reference parameter, and a graphical user interface.
10. The method of claim 1 , wherein the user-defined region is specified by a user via a graphical user interface.
The user-defined region on the class reference image, which is used to create the class reference data record, is specified by a user via a graphical user interface. The user visually selects the region of interest. This builds upon the method where a class reference image is created from images of known banknotes, a user-defined region selected in the image, and a class reference parameter calculated as a ratio of pixel values from infrared-absorbing and transparent regions. Each pixel in the class reference image is either the average or median of corresponding pixels from the reference images and a variable tolerance range is assigned to the class reference parameter. The class reference image is displayed via a GUI.
11. An apparatus for processing value documents, in particular bank notes, wherein the apparatus comprises: a processor; a memory; a hardware sensor for generating an image of a bank note including at least a lens and a light sensor; a control and evaluation device, wherein the control and evaluation device is configured to carry out following steps: creating a class reference image by means of a plurality of images of already classified value documents of same class, the plurality of reference images being created using an imaging optical sensor; and creating, by using the processor, a class reference data record having at least one class reference parameter with the aid of a user-defined region having at least one optical feature within the class reference image, the class reference parameter being coupled to at least one quantitatively determined property of the user-defined region, the class reference parameter being created by forming a ratio of pixel values from at least two different locations on a value document, the ratio being formed from a pixel value averaged over a specified region of an infrared-absorbing section of the value document to a pixel value averaged over a specified region of an infrared-transparent section of the value document; wherein the pixel values of a respective pixel of the class reference image are at least one of the average pixel values or median pixel values of the corresponding pixels of the plurality of images of already classified value documents of the same class, wherein each class reference parameter has assigned thereto a tolerance range, the tolerance range for each class reference parameter being variable and being dependent on the quantitatively determined property of the user-defined region; and wherein the class reference image of a class of value documents is represented, for analysis, on a display unit by means of a graphical user interface.
An apparatus for processing banknotes consists of a processor, memory, a hardware sensor with a lens and light sensor to create an image of the banknote, and a control/evaluation device. This device creates a class reference image from multiple images of known banknotes using the optical sensor. A class reference data record is then created, based on a user-defined region within the class reference image, and it contains at least one class reference parameter. This parameter is linked to a quantitative property of the region, and it’s calculated as the ratio between pixel values from infrared-absorbing and infrared-transparent sections. Each pixel of the class reference image uses average or median values. Each class reference parameter has a variable tolerance range, and the class reference image is displayed via a graphical user interface.
12. The apparatus according to claim 11 , wherein the control and evaluation device is further configured to carry out the following further steps: determining at least one quantitative property of a value document to be classified; and classifying the value document to be classified on the basis of a comparison of the quantitative property of the value document to be classified with the class reference parameter.
The banknote processing apparatus can determine at least one quantitative property of a banknote to be classified, and then classifies the banknote by comparing this quantitative property to the class reference parameter. This classification functionality builds upon the core apparatus with a processor, memory, and sensor that creates a class reference image from multiple banknote images, defines a user-defined region with optical features, calculates a class reference parameter based on the ratio of pixel values from infrared regions, and displays a class reference image on a GUI with variable tolerance ranges.
13. The apparatus according to claim 11 , wherein the apparatus further comprises a computer having a display unit connected with the control and evaluation device, wherein there is implemented on the computer a graphical user interface allowing a user to participate in the step of creating a class reference data record having at least one class reference parameter with the aid of the class reference image.
The banknote processing apparatus includes a computer with a display unit connected to the control and evaluation device, with a graphical user interface that allows the user to participate in creating the class reference data record using the class reference image. This allows user input in parameter creation. This extends the core apparatus including a processor, memory, sensor, and control/evaluation device that creates a class reference image, defines a user-defined region with optical features, calculates a class reference parameter based on pixel values from infrared regions, displays the image on a GUI, and includes variable tolerance ranges.
14. A computer-readable hardware storage device comprising a memory having stored thereon a program for determining a class reference data record for the classification of value documents, comprising: creating a class reference image by means of a plurality of reference images of already classified value documents of same class, the plurality of reference images being created using an imaging optical sensor; and creating, by using a processor, a class reference data record having at least one class reference parameter with the aid of a user-defined region having at least one optical feature within the class reference image, the class reference parameter being coupled to at least one quantitatively determined property of the user-defined region, the class reference parameter being created by forming a ratio of pixel values from at least two different locations on a value document, the ratio being formed from a pixel value averaged over a specified region of an infrared-absorbing section of the value document to a pixel value averaged over a specified region of an infrared-transparent section of the value document; wherein the pixel values of a respective pixel of the class reference image are at least one of the average pixel values or median pixel values of the corresponding pixels of the plurality of reference images of already classified value documents of the same class, wherein each class reference parameter has assigned thereto a tolerance range, the tolerance range for each class reference parameter being variable and being dependent on the quantitatively determined property of the user-defined region; and wherein the class reference image of a class of value documents is represented, for analysis, on a display unit by means of a graphical user interface.
A computer-readable hardware storage device contains a program that determines a class reference data record for classifying banknotes. The program creates a class reference image using multiple reference images of already classified banknotes. Then, it creates a class reference data record, with at least one class reference parameter, based on a user-defined region within the class reference image. The parameter relates to the quantitative property of the region, created by forming a ratio of pixel values from infrared-absorbing and transparent areas. Each pixel of the class reference image utilizes average or median pixel values. Every class reference parameter is assigned a variable tolerance range, and the class reference image is displayed for analysis by a graphical user interface.
15. The computer-readable hardware storage device of claim 14 , wherein the computer program is stored on a data carrier.
The computer program from the hardware storage device, which determines a class reference data record, is stored on a data carrier. This program creates a class reference image, extracts a user-defined region with optical features, computes class reference parameters related to quantitative properties of the region based on ratios of pixel values from infrared regions, and includes variable tolerance ranges; The class reference image is displayed with average/median values.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
December 14, 2011
June 27, 2017
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.